1
$\begingroup$

Here’s the idea: I want to create a biological info-storing material, probably much like RNA, that—like noncoding RNA does—is able to mediate and enact epigenetic changes to the genetic code/material of an organism according to its instructions.

Here’s the problem: In an interspecies and intergalactic setting, I want this RNA-like material (we’ll call it Substance X) to be able to function across multiple different systems of nucleic acids and genetic materials, not just our DNA/RNA chemistry but also the potential analogues to it. Now because it’s just for epigenetic changes (only adding chemical “tags” to genes in order to affect how they’re read and expressed), Substance X does not have to be fully integrated into an organism; it’s never actually read or replicated, so its relationship to an organism’s genetic material is very “one-way”.

Here’s the potential solution: Since we’re already working with hypotheticals and imaginary biochemical materials here, I’ve considered further creating some sort of chemical group/protein/etc that Substance X uses (let’s call it Component Y). Component Y is able to identify which segments of genetic material prompt which proteins/traits/functions across different genetic code systems, in order to be able to target the right gene section for its job and make the same overall effect across different types of XNA. (Maybe it’s akin to what I’ve read about transcription factor proteins, in how it “reads” DNA to find the right sequence for its job.)

In short, rather than just being inherently built for a certain target, somehow it searches through or “observes” genetic code once it’s in an organism’s genetic material, and figures out where to target by doing so.

And here’s the question about it: Chemicals, proteins, or other molecules like that do have all sorts of behaviors and protocols including “searching” abilities (in fact the aforementioned transcription factors, and findings about their ability to “read” DNA, is maybe the closest real comparison I’ve found)…but they aren’t conscious, thinking entities, and the problem I’m giving them here is about dealing with all sorts of different systems and probably can’t be solved with just “look for this pattern” programming.

Is this sort of gene-code-reading function something that a hypothetical biomolecule/chemical (or group of them) could pull off, even within a wide range of genetic system and “language” types? Or is that too much of a demand to try and let something like our “Component Y” carry?

$\endgroup$
3
  • 2
    $\begingroup$ Fair warning: it's usually impossible to simply explain "how could X work?" In fact, the answer to "would that be possible" is 99.9999% "no." What's more likely to happen is that respondents will show (if any exist) terrestrial examples of similar behavior that you could then use to model your imaginary world's rules. A recent upsurge in questions asking us to make their fiction factual has raised a bit of frustration because, per the help center, our goal is to help you build the rules of an imaginary world, not bring fiction into the Real World. Thus, can you clarify your expectations? $\endgroup$
    – JBH
    Jun 20, 2023 at 0:32
  • $\begingroup$ @JBH that makes sense, my bad! I’ll see if I can edit the question to narrow it down and clarify my thoughts some $\endgroup$
    – inkwell87
    Jun 20, 2023 at 8:29
  • 1
    $\begingroup$ Thanks, but now we have a second problem. How is this question not a duplicate of your earlier question, Could a (semi-)universal information-storing molecule exist?? Quoting from that earlier question, "Is there some sort of (hypothetical) biological substance I could use as an information-carrier, akin to DNA/RNA but compatible with all sorts of biochemistries rather than just one system of nucleic acids?" $\endgroup$
    – JBH
    Jun 20, 2023 at 14:29

4 Answers 4

2
$\begingroup$

The short answer is: No.

You might be able to work something out that will work with terrestrial life that uses DNA. DNA and its behavior is a known quantity now.

Figuring out something that will also work with the genetic material of a single alien biome would be orders of magnitude more difficult, since aliens may not use DNA as a genetic material at all, and it would require as much knowledge about the alien genetic material and its workings as we have about DNA. What makes it so much more difficult is that it would have to work on both types of genetic material.

Something that would work on three or more types of alien genetic material is probably a great big fat 'Hell, no!'. You would have to know how they all work, and incorporate all that data in one 'molecule'. To get anywhere near a solution to this for just two or three different types of genetic material, you'd need a pretty substantial nanite packed full of processing power and data... hardly a single molecule.

The problem is that while 'genetic material' is probably a biological universal, the exact means that each independently-occurring and evolved genetic system uses is very likely a privative, which is to say that each biosphere will have its own unique genetic mechanism. Maybe more than one. And none of them are going to work the same way as any of the others.

To give examples, terrestrial DNA uses three base pairs to encode to one amino acid. Alien genetic material may not even be DNA, and it may not even encode to amino acids... and whatever it encodes to, it may require a different number of bases to encode it. The genetic replication mechanisms will be different. The means by which replication is initiated will be different.

And finally, if you don't know how an alien biosphere's genetic inheritance works, your 'molecule'/nanite is going to have to do all that research for itself before it can even start doing what you want. You're definitely pissing into the wind trying to do that.

The best you're going to be able to do is design one nanite/nanite suite for each biosphere that you're intimately familiar with. If you're not intimately familiar with a biosphere's genetic mechanisms, then a mere nanite, no matter how smart, isn't going to be of much use any time soon.

$\endgroup$
5
  • $\begingroup$ :) Just read your A, and your comment on mine. Funnily, i think we are on the same page - you explaining from the No-because direction, me from the Yeah-but-no direction :) $\endgroup$
    – bukwyrm
    Jun 21, 2023 at 13:19
  • $\begingroup$ So the “why” I’m seeing here is that the variety of different (interplanetary) species’ genetics systems, even if we’re just talking “multiple” and not all or infinite, is too much different info for this thing to contain. I kinda suspected that problem to start with—hence the whole approach of “whatever system this is must search for its target area and won’t just know naturally since it’s in a foreign genetic type”, to somehow allow it to work case-by-case instead of having to contain EVERY possible “case” within it going in. I might have to push that further to make this (continued) $\endgroup$
    – inkwell87
    Jun 21, 2023 at 17:43
  • $\begingroup$ concept work somewhat more plausibly (still very much within the realm of hypothetical sci fi though of course). Question, does it help at all if this substance only has the purpose of making one specific change to one specific genetic trait, rather than having to operate with knowledge of what ALL of EVERY species’ genetic code means? Or alternatively would it be easier to make this work by having this substance’s effect be put in place on the protein level—suppressing proteins produced by genes to mediate changes—instead of having to navigate through different genetic code systems? $\endgroup$
    – inkwell87
    Jun 21, 2023 at 17:49
  • $\begingroup$ @inkwell87 Evolved systems tend to be a tangled mess, and it takes an incredible amount of effort to understand what's going on. You almost have to understand everything in order to change how one thing works, or else risk serious unintended consequences. Also, you don't even know if a particular biochemistry even uses proteins. To make this work, you're heading into the realm of magic. $\endgroup$
    – Monty Wild
    Jun 22, 2023 at 2:59
  • $\begingroup$ "Question, does it help at all if this substance only has the purpose of making one specific change to one specific genetic trait": it's faulty to assume that there will be any correspondence at all in the genetics of two organisms relating to a given trait, let alone one specific change required for it. Genomes aren't CAD models where you just need to learn the file format. It might be a direct consequence of a single gene in one species, and a result of a complex system of interactions between multiple genes controlling other systems in another. $\endgroup$ Jul 21, 2023 at 14:45
0
$\begingroup$

You can have something like that, as long as you keep your expectation for 'read' low.

Thing is, we do not 'read' DNA either. Using a lot of computing power, and still not completely flawlessly, we can now directly go 'code X -> protein Y'. In most cases that is not what the gene 'does', though. We take organisms, fiddle around with their genetic code, look for what breaks. Correlate change with breakage, fiddle around some more, note breakage, fiddle, note, etc. - then we have some picture of what that bit 'does'. It could be that the protein that is being encoded is directly used (as a weapon, for instance) and only used directly, but mostly it is just part of a causality chain involving other molecules. So we cannot do a cold reading: "ah, ATAGACTTGATTACA - that means the carrier will be a thoughtful lover" - and reversely, we cannot have a mechanism that gets the info : 'make everyone a thoughtful lover' and goes out and does it.

That said, what we can do (... at some point in the future) is use the info we have, from fiddling and noting, on what (epi)genetic changes will keep people from being thoughtful lovers, make a molecule or set of molecules that will effect the correct changes (and ONLY make any changes in humans, not in the same code that does , e.g, waste management in mosses) , make a system that ensures those molecules reach the cells' nucleus, then click 'print'.

Doing that for different mechanisms of inheritance at the same time is just adding complexity.

We might conceivably, some day, be able to do the 'fiddle, note' part in silico, i.e. do a whole-organism simulation, but we'd still need the complete genetic code to begin with, i.e. a molecule, or molecules, bobbing alongside the DNA in the nucleus, interacting with only a few basepairs at a time, CANNOT, ever, come in with the trait-info, and target the code responsible.

$\endgroup$
8
  • $\begingroup$ This would apply only to terrestrial biology. This doesn't even begin to consider the consequences of alien biology. $\endgroup$
    – Monty Wild
    Jun 21, 2023 at 13:08
  • $\begingroup$ @MontyWild - what could i write there? we do not know what alien biology is out there. Let's imagine it would for some reason involve loads of lead and mercury and store info in doping gradients in crystals - that would of course not help make things go more smoothly, but the general principle applies. $\endgroup$
    – bukwyrm
    Jun 21, 2023 at 13:14
  • $\begingroup$ Exactly my point. How do you design a system that operates on all of the possible ways that an organism might store genetic data, when you don't even know all the possible ways? That's what the OP is asking. $\endgroup$
    – Monty Wild
    Jun 21, 2023 at 13:17
  • $\begingroup$ @MontyWild - i understood the Q such that WE do not know about the alien biochem, but the people developing that epigenetics solution do. $\endgroup$
    – bukwyrm
    Jun 21, 2023 at 13:21
  • 1
    $\begingroup$ As I explained in my answer, you can do it for one system, maybe for two, but it'd be very much harder. For three or more, practically impossible. Each extra system is adding a vast amount of complexity... and that's if you know exactly what you're doing. If you don't know, then no way. $\endgroup$
    – Monty Wild
    Jun 21, 2023 at 14:03
0
$\begingroup$

Biochemistry is too particular for this to work

Things like viruses work by hijacking the machinery of your cells by using the same kind of genetic code and biochemical components as the cell, but if Substance X is not even based on the same kind of biochemistry as an alien life form, then it is nothing more than a foreign contaminate. If for example, you inject this stuff into a lifeform that does not use nitrogen as a major building block, there will not be enough nitrogen in the cellular fluid to synthesize amino acids; so, it just spits out piles of random, broken up hydrocarbons that will likely just poison and kill the cell.

Adding Component Y is even less reasonable because reading genetic code requires the ability to chemically bind yourself to each nucleotide in the genetic code. If your host organism has a single kind of nucleotide that Component Y does not have a one and only one specific way to chemically bind too, then it can not read that nucleotide. It would be like trying to compile a program if a quarter of the letters in the alphabet caused fatal errors.

The closest you can plausibly get

Substance X and Component Y are not unique chemicals, rather they are classes of chemicals. Each organism has a specific Substance X that is custom designed to have an epigenetic effect on the organism, and you need to create a unique Component Y for each kind of translation you want to allow. So, if you want to transfer an epigenetic effect from species a to species b, you must extract Substance X.a from species a, add Component Y.ab in a solution with all of the necessary base elements to transcribe the information into Substance X.b, then inject Substance X.b into species b.

This method will never work between arbitrary life forms, but any alien lifeform you have time to study in enough detail, you could develop a new Component Y for.

enter image description here

$\endgroup$
-1
$\begingroup$

CRISPR-CAS variation

CRISPR-CAS does 90% of what you want. A bacteria has a very basic defence mechanism. If it survives a virus attack, it will make a protein with part of the virus DNA. The moment this exact sequence is detected, the protein will cut out that part of the DNA and replace it with some garbage DNA. This will prevent the virus from functioning and thus keep the bacteria safe. We've found out we can change this process, making it target a specific DNA strain we want and replacing it with something we want to put in there. It is on the go cutting out something we don't want and pasting something we do want.

From here it is a question what is the least far fetched from here. I assume much DNA is ordered in recognisable patterns, making most DNA sequences identical. We share much DNA with a banana for a reason. That also means much of the, for example, skin DNA is probably (near) identical. Only a few bits and pieces of the DNA decide on the properties that truly are different, like colour or it's general strength. But compare one skin cell to that of some creature on the opposite side of the globe and you'll see that much of the DNA is similar if not identical.

We can use this for a plausible scenario. Instead of cutting and pasting only a very specific code, it recognises the start and end of what you want to target. It recognises skin cells start/end/middle where it is practically always identical, and then adapts the sequences near it to get the desired effect.

Using viruses and bacteria

We can extend this idea even further. If you need something to make 'decisions', why not use a symbiote? A bacteria or a virus in a cell can produce a ton of enzymes and reproduce themselves. The resulting proteins and chemicals can signal what is needed by temporarily attaching to some DNA/RNA cluster, which the bacteria or viruses adapt to. They produce the corresponding proteins to cut and paste for you. As this is a symbiotic relationship, like the mitochondria, both organisms can improve and survive.

$\endgroup$
2
  • $\begingroup$ Sure... for terrestrial biology. What about some alien biome that doesn't even use DNA? $\endgroup$
    – Monty Wild
    Jun 21, 2023 at 13:08
  • $\begingroup$ @MontyWild as we don't know what 'analogues' an alien uses, it seems fine to me to have proteins that can cut and paste whatever. As far as I know the proteins can work on anything, but these are certainly great for DNA. The question also seems to focus primarily on DNA, only sometimes referring to analogues. $\endgroup$
    – Trioxidane
    Jun 21, 2023 at 18:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .